In virtually all domains, attempts at classification have been made. Many of these attempts started with a one-dimensional system based on a variant of the Dewey decimal systems. These systems were augmented with facetted keywords, creating ex-post catalogues of words that were created based on the existing system. While these efforts were normal knowledge processes created on an iterative basis that helped characterize specific domains, they were not rules-based. As a result, there are significant limitations to how data structured in this manner can be used and analyzed.